Chapter 2: Data-Driven Problem Solving

The definition that will be provided for data mining will underline the main reason behind referring to the set of activities that are presented in this book as data-driven modeling rather than data mining. It is true that we are using data mining (among other tools) to perform the analyses (building reservoir models) that are presented here, but, as will become more and more clear, our objective is not merely to identify the utility of yet another newly popularized tool in the oil industry (a task that by itself may have merit and may be treated as an academic exercise), but rather to offer a solution and a tool that can be used by professionals in our industry today. Ever since its introduction as a discipline in the mid-1990s, Data Science has been used as a synonym for applied statistics. Today, Data Science is used in multiple disciplines and is enjoying immense popularity. Application of Data Science to physics-based (such as the oil and gas industry) vs. nonphysics-based disciplines has been the cause of much confusion. Such distinctions surface once Data Science is applied to serious industrial applications rather than to simple academic problems.